Back in November, I sent out a query on experience with Gemini-Mercury upgrade
experiences. I am attaching anonymized responses for the interest of the group.
Summary: Most people were pleased overall, especially with simplicity, with
software and with gradient shimming.
Major concerns:
heavy use of the 9-sample changer in a walk-up environment
Varian 4-nuc probes (especially if tuning required)
use for solids work
I had a demo in PA one week ago & was pleasantly surprised by the machines
capabilities, especially with the ChemPack integration. We will probably be
ordering one in the near future. (In biotech, nothing except a stock-drop is
definite until it happens.)
Thanks again to all who responded.
Patrick
The original query:
Fellow Magnet Drivers,
The wheels are starting to fall off of my 8+ year old Gemini 200. Failure is
following on problem which is following on catastrophe, resulting in lots of
instrument down-time & lots of my repair time. The only good thing to come out
of this is support at budget time for an upgrade. We are currently planning on
a Mercury VX, but would like some opinions before we plunge in with our capital
dollars.
1- For those who have made this jump, are you pleased with the transition?
Anything unexpected I should look out for?
2- Has anyone made an ASM-100 work with the Mercury / Mercury VX? Varian is
not professing support for their older-model sample changer.
3- Are lab managers pleased with the new 9-sample carousel sample changer?
Have the problems been ironed out?
4- Is anyone running gradient-shimming in a walk-up environment? How is it
working? How about with automation?
I would appreciate any experience that could be relayed which might be relevant
to this set of questions. I'll put together a summary with identifiers deleted
for any who request. I've looked over the results of "carousel" & "mercury"
searches on the AMMRL at UN-L, and have a good sense of what has been said
there. Any updates from those posts would be welcome.
*****************************************************************
Responses:
1.
We have an ASM on a 9 year old Gemini which has already been upgraded to a
Gemini 2000. The ASM works fine. Don't know about Mercurys however.
We have a 9-sample autosampler on Unity system and it has been great since
day 1. I highly recommend them. No major problems provided the air pressure
supply for sample lift an carousel rotation is stable.
2.
This past spring we upgraded our 10 year old GE OMEGA spectrometers to
Varian MercuryVx consoles with the new Zymark sample changer. Both systems are
working fine and are used in a walkup mode with deuterium gradient shimming.
The spectra are generally much better than we got previously using the old GE
auto shim program. I think you will not be disappointed if you do the
upgrade. By the way we kept our old 300 MHz magnets but had the bore tube
changed from 44 mm to 54 mm to accommodate the standard Varian probes.
3.
A Mercury is a big step up from where you are now. Varian made big
improvements with the jump to Mercury. There is almost nothing it
can't do especially compared with your Gemini. I have (or have had)
both and I also have Inova spectrometers. If you want specifics then
please ask and I'll try to help you out. Basically, the VX is very
stable, has extremely clean RF, quiet, small package (plugs into the
wall!), as you know comes with several options including wavegen if
you need them for something. My user group prefers it to the Inova
for most things.
I have two 9-sample carousels. Once you get them working they do
just fine; they are robust and don't have any bad failure modes that
I've encountered. However, I *do not* use them in a multi-user,
"walk-up" lab. They are not large enough for this purpose, they are
challenging for users to work with compared with the ASM or the new
robots, and you *will* introduce failures by going down that path.
Instead, I (or other users) load up a carousel with a project and
then have nine samples under remote control. That is a very nice
feature. The rest of the time I remove the carousel and we operate
the instrument quite efficiently in "walk-up" mode.
Gradient shimming is fantastic. Krish Krishnamurthy has put a lot of
effort into a product called (I believe) Chem Pack. One of the
latest features (which may not be part of their commercial release
yet) is the fashion by which Varian "acquires" the lock (with a
pulse/acquire/set-lock through the acquisition channel) and and then
gradient shims. This is quite robust and very fast. Steve Cheatham
tells me that fifteen samples per hour (mixed solvents) is not
unreasonable...with robot-tweaking and compromise on the acquisition
you may approach twenty samples per hour.
4.
We recently upgraded one of our QE 300's to the Mercury VX console.
Ofcourse this was a giant leap in experimental capabillities and we are
very pleased with its performance. We have a multi user environment and the
short learning curve is a giant plus for me. Also the gradient shimming is
a real life save for those not adept at shimmimg. I do find though that the
gradient shimming works best with just the right amount of solvent and best
results are really obtained by mapping each solvent seperately,
unfortunately the software by default will always use the last map
generated. But overall I would give the Mercury VX a big plus.
5.
We have the 4-nuc instrument. Actually, we have the older Mercury
(not the VX). I intend to upgrade to the VX "soon" since there are
some additional features there not found on the first generation cpu.
We demoed the broadband and, frankly, saw no difference on our
instrument.
Where you will get into the most trouble with the carousel is if you
attempt to use it in concert with traditional walk-up. That just
doesn't work well. The mechanics of it are nasty and the context
switching (in and out of automation) are also nasty. Frankly, I
think that is a hardware limitation.
Now, it is possible to run walk-up for awhile (say, during the busy
part of the day) and then run the carousel for awhile (say overnight
or on the weekend). That works well for us. Indeed, I really like
it because I can run the system from home, play with my kids, etc.
and still say I'm working without ever hitting the U.
One possible limitation to all of this will be the range of sample
types that you place in a single run. Various solvents will detune
the probe differently (of course) and this can have dramatic
consequences for your work, even simple 13C{1H} can fail under some
conditions and more sensitive experiments will have even more dramtic
problems. I believe Varian is making headway in this regard with
their probe design but it isn't a done deal. Nalorac claims to have
a probe style that takes care of "tuning" issues but I don't know
anything more than that.
Thanks very much,
Patrick
Patrick Wheeler
Isis Pharmaceuticals pwheeler_at_isisph.com
2280 Faraday Avenue phone: 760 603 2409
Carlsbad, CA 92008 fax: 760 929 0036
Received on Tue Mar 14 2000 - 19:12:57 MST